Implementation and validation of the total organic carbon determination method in liquid matrices from the pulp and paper industry

• Autor(a) principal: Sacramento, Margarida Menezes Amador
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/28212
• topic: Validation; Total organic carbon (TOC); COD/TOC relationship; Effluents; Pulp and paper (P&P) industry

The pulp and paper (P&P) industry is one of the most important industries in the world and also one of the largest consumers of water. High water consumption implies the production of a high volume of effluents, being one of its characteristics the high concentration of organic compounds, which can have an environmental impact in the receiving aquatic systems, due to the decrease in the oxygen concentration. Chemical oxygen demand (COD) is widely used for monitoring the organic matter present in wastewater but, due to the production of hazardous wastes during the analysis, has increased the interest in replacing this method by total organic carbon (TOC). Following the acquisition of a TOC analyser by the RAIZ laboratory, the present study consisted in validating the direct method (NPOC) for TOC analysis. The evaluated parameters were the working range, linearity, sensitivity, limit of quantification (LQ) and precision. The factory working range, [0.3–3.0] μg was tested, having been the homoscedasticity condition of the data confirmed. The calibration curve, of the NPOC peak area as a function of NPOC weight, showed good linearity, proven statistically and by the correlation coefficient (r=0.99989). Sensitivity, assessed by the slope of the calibration curve, proved to be satisfactory. Also, the average slope deviation (4.9 %) from the factory slope was within the established acceptance criteria, i.e., 5 %. The quantification limit (LQ – 0.24 μg) proved to be appropriate for the method. The repeatability and intermediate precision (IP) were assessed for untreated and treated effluent. The former was acceptable for both matrices, with both CVr values below 10 %, 4.3 %, for untreated influent, and 8.9 %, for treated effluent. The repeatability limits were also defined for untreated (Δr=0.093 μg) and treated (Δr=0.071 μg) effluents. The IP was evaluated by range control charts, which showed that the process was under control for both types of samples. Reproducibility (CVR=8.2 %) was assessed through the participation in an interlaboratory test. The TOC consensus value of the wastewater sample was 91.9 mg/L and a Z-score of 0.12 was attributed to RAIZ. The uncertainty of the NPOC method was also calculated for both untreated (U=28.3 %) and treated effluent (U=22.3 %) matrices. Also, a study was carried out to establish COD:NPOC correlation factors for untreated and treated effluents from the P&P industry. The y=bx+a model was used to establish equations for both matrices, and b values of 3.02 and 2.36 were obtained, respectively. The validation of the difference method (DM) for TOC analysis was also started. However, the factory calibration proved not to be in effect and a new calibration was performed and properly validated by the analysis of a certified reference material. Finally, due to the total nitrogen (TNb) analysis capability of the TOC analyser purchased by RAIZ, the equipment was also calibrated for the surface waters matrix analysis